Introduction

The complete structure of an unknown glycoside comprises multiple levels of information, including the molecular formula, aglycone structure, types and numbers of sugars, linkage position between the sugar and aglycone, sugar-chain linkage sequence, and anomeric configurations. High-resolution mass spectrometry provides information on elemental composition, chemical hydrolysis and monosaccharide analysis determine the constituent units, and nuclear magnetic resonance spectroscopy establishes atomic connectivities and provides stereochemical information. The final structure is defined by the combined constraints imposed by these different experimental results.

Amakura et al. isolated a new flavonoid glycoside from an extract of Ephedra herb. Structural investigation of this compound involved high-resolution electrospray ionization mass spectrometry, acid hydrolysis, analysis of the absolute configurations of the monosaccharides, and one- and two-dimensional nuclear magnetic resonance spectroscopy. The aglycone, the two constituent sugars, the sugar–aglycone linkage position, the sugar–sugar linkage position, and the two anomeric configurations in the final structure were each supported by corresponding experimental evidence [1].

1 HRMS and MS/MS Provide Molecular Composition and Fragmentation Information

1.1 From the Observed Ion to the Molecular Formula

Mass spectrometry measures the mass-to-charge ratio (m/z) of charged particles. During electrospray ionization, natural products can form protonated ions, deprotonated ions, and metal-ion adducts, such as [M+H]⁺, [M−H]⁻, and [M+Na]⁺. The ion form determines the relationship between the observed m/z value and the mass of the neutral molecule.

Amakura et al. observed the following ion for the unknown flavonoid glycoside by high-resolution electrospray ionization mass spectrometry (HR-ESI-MS) in negative-ion mode:

m/z 609.1466 [M−H]⁻

The calculated theoretical value reported by the authors was:

m/z 609.1461

corresponding to the neutral molecular formula:

C₂₇H₃₀O₁₆

The ¹³C nuclear magnetic resonance spectrum also showed carbon signals consistent with 27 carbon atoms, supporting this molecular formula [1].

High-resolution accurate-mass measurements can substantially narrow the range of possible elemental compositions. Nevertheless, multiple candidate elemental compositions may still fall within the same accurate-mass range. Isotopic distributions, permitted element types, elemental ratios, valence relationships, and other independent spectroscopic information can be used to further filter candidate molecular formulas. Systematic studies by Kind and Fiehn showed that even at mass errors below 1 ppm, accurate mass alone may still yield multiple elemental compositions that satisfy the specified criteria [2,3].

The molecular formula C₂₇H₃₀O₁₆ describes the elemental composition of the entire molecule. The aglycone skeleton, positions of hydroxy substituents, specific sugar identities, sugar-chain linkage pattern, and stereochemical configuration constitute additional levels of structural information.

1.2 Information on Sugar Residues from MS/MS

Tandem mass spectrometry (MS/MS) selects a specific precursor ion and induces its fragmentation, providing local structural information from the masses and elemental compositions of the resulting product ions.

Flavonoid O-glycosides commonly undergo cleavage of glycosidic linkages during collision-induced dissociation (CID). For example, when consistent with the structural context and fragmentation pathway, a neutral loss of approximately 162 Da can support the presence of a hexosyl residue with the elemental composition C₆H₁₀O₅.

When aldohexoses such as glucose, galactose, and mannose form glycosides, their glycosyl residues contribute the same nominal mass. Therefore, a mass loss of 162 Da provides information at the level of a hexosyl residue; the specific monosaccharide identity must be further established by monosaccharide analysis, comparison with reference standards, or other independent structural evidence.

Disaccharide and polysaccharide chains may also produce sequential losses of sugar residues, losses of internal monosaccharide residues, and glycosidic-bond cleavage products with different relative abundances. In a study of four pairs of isomeric flavonoid O-diglycosides, Ma et al. found that losses of internal monosaccharide residues and Y-type product ions formed by glycosidic-bond cleavage could reflect the type of sugar-chain linkage [4].

1.3 Interglycosidic Linkage Pattern Affects MS/MS Fragmentation

For disaccharide chains composed of rhamnose and glucose, two common linkage patterns are:

α-L-rhamnopyranosyl-(1→2)-D-glucosyl

α-L-rhamnopyranosyl-(1→6)-D-glucosyl

The former corresponds to the neohesperidose structure, whereas the latter corresponds to the rutinose structure.

In the flavonoid O-diglycosides they studied, Ma et al. found that losses of internal monosaccharide residues and the relative abundances of Y-type ions could distinguish isomers containing 1→2 and 1→6 linkages [4]. Cuyckens et al. investigated similar flavonoid O-diglycosides by positive- and negative-ion nano-electrospray ionization ion-trap mass spectrometry and likewise observed different fragmentation characteristics for the two interglycosidic linkage patterns; ionization mode and aglycone structure affected the relative abundances of the product ions [5].

Mass differences between MS/MS fragments can also correspond to specific bond-cleavage processes. In a study of flavonoid O-diglycosides from the genus Citrus, Shi et al. found that the [M−H−120]⁻ ion of naringin had previously been assigned to sugar-chain fragmentation, whereas accurate-mass data from orthogonal time-of-flight mass spectrometry showed that this product originated from retro-Diels–Alder (RDA) cleavage of the flavonoid C ring [6].

The accurate masses and elemental compositions of product ions, together with the precursor structure and fragmentation reactions, collectively constrain the origin of individual fragments. MS/MS can provide information on sugar-residue composition, partial sugar-chain sequence, and linkage type in specific systems; the specific carbon positions involved in the linkage are established by structural evidence such as nuclear magnetic resonance spectroscopy.

2 Hydrolysis and Monosaccharide Analysis Determine the Constituent Units

2.1 Determination of the Aglycone by Acid Hydrolysis

After cleavage of the glycosidic bond, the released non-sugar moiety can be independently analyzed by chromatographic and spectroscopic methods. Amakura et al. treated 1.5 mg of the unknown glycoside with 0.5 mL of 0.1 mol/L hydrochloric acid and heated the mixture in a boiling-water bath for 5 h. After cooling, the reaction mixture was processed by C18 solid-phase extraction, and the methanol-eluted fraction was analyzed by reversed-phase high-performance liquid chromatography (HPLC).

Herbacetin with a retention time of 27.3 min was detected among the hydrolysis products and confirmed by co-chromatography with an authentic herbacetin reference standard [1].

This hydrolysis experiment established the compositional relationship between the unknown glycoside and herbacetin:

Unknown glycoside → acid hydrolysis → herbacetin

Herbacetin was therefore identified as the aglycone of the glycoside.

The aglycone obtained by acid hydrolysis corresponds to the non-sugar moiety ultimately detected under the specified hydrolysis conditions. Natural products containing acid-sensitive structural features may undergo further reactions during hydrolysis. The specific hydrolysis conditions and the aglycone reference standard therefore together define the experimental context for aglycone identification.

2.2 Monosaccharide Identity and D/L Configurational Series

Hydrolysis of a glycoside also releases its constituent sugars. The D- and L-forms of the same monosaccharide are enantiomers. After free sugars react with a chiral reagent to form diastereomeric derivatives, these derivatives can be separated by conventional reversed-phase chromatography.

The reversed-phase HPLC method developed by Tanaka et al. uses L-cysteine methyl ester and o-tolyl isothiocyanate to derivatize aldoses. The resulting diastereomeric derivatives can be separated on a C18 reversed-phase column, and multiple D- and L-aldoses can be distinguished by ultraviolet detection [7]. The original study validated aldoses including glucose, mannose, galactose, rhamnose, fucose, arabinose, and xylose; ketoses such as fructose did not yield analytical peaks suitable for this method under the experimental conditions used.

Amakura et al. applied this method to analyze the constituent sugars of the unknown glycoside. A 1.0 mg sample was hydrolyzed by heating with 0.2 mL of 0.5 mol/L hydrochloric acid. After neutralization with a strongly basic anion-exchange resin and evaporation to dryness, the residue was reacted with L-cysteine methyl ester hydrochloride in pyridine at 60 °C for 1 h. o-Tolyl isothiocyanate was then added, and the reaction was continued at 60 °C for another 1 h. The reaction mixture was analyzed by reversed-phase HPLC, and the resulting chromatographic peaks matched those of D-glucose and L-rhamnose reference derivatives prepared using the same procedure [1].

The constituent sugars were identified as:

D-glucose

L-rhamnose

D/L denotes the configurational series to which a constituent sugar belongs. Once the monosaccharide identity has been established, determination of its D/L form identifies the absolute configuration of the corresponding enantiomer. The α/β configuration at the anomeric carbon in the intact glycoside represents a separate level of stereochemical information and is determined from nuclear magnetic resonance data for the intact molecule.

2.3 Constituent Units Obtained by Hydrolysis

Acid hydrolysis and monosaccharide analysis identified three constituent units:

Herbacetin

D-glucose

L-rhamnose

The sugar–aglycone linkage position, the linkage position between rhamnose and glucose, and the anomeric configurations of the two sugar residues in the intact glycoside are retained in the nuclear magnetic resonance signals and internuclear correlations of the unhydrolyzed molecule.

3 Nuclear Magnetic Resonance Establishes the Structural Units of the Aglycone and Sugars

3.1 Nuclear Magnetic Resonance Signals of the Herbacetin Skeleton

Proton nuclear magnetic resonance (¹H NMR) and carbon-13 nuclear magnetic resonance (¹³C NMR) reflect the chemical environments of hydrogen and carbon atoms within a molecule. Correlation spectroscopy (COSY) establishes coupling relationships between protons within the same spin system, whereas heteronuclear single quantum coherence (HSQC) correlates protons with the carbon atoms to which they are directly attached.

The ¹H NMR spectrum of the unknown glycoside contained two sets of aromatic doublets, each integrating for two protons:

δH 6.90, 2H, d, J = 9.0 Hz

δH 8.20, 2H, d, J = 9.0 Hz

An additional signal was observed at:

δH 6.62, 1H, s

The two sets of doublets form an A₂B₂-type proton system commonly observed for a para-disubstituted benzene ring. Fifteen sp² carbon signals attributable to the flavonoid skeleton could be assigned in the ¹³C NMR spectrum. On the basis of these spectroscopic features, the original study proposed a herbacetin-type flavonoid skeleton, and the identification of herbacetin after acid hydrolysis by comparison with a reference standard further confirmed the identity of the aglycone [1].

3.2 Nuclear Magnetic Resonance Signals of the Two Sugar Residues

Two sugar anomeric proton signals were observed in the ¹H NMR spectrum:

δH 5.20, 1H, d, J = 7.5 Hz

δH 5.29, 1H, d, J = 1.5 Hz

Also observed was:

δH 1.90, 3H, d, J = 6.0 Hz

together with largely overlapping sugar-region proton signals in the δH 3.3–3.9 range. The methyl doublet at δH 1.90 is consistent with the structural characteristics of the C-6 methyl group of rhamnose [1].

Key sugar-carbon signals in the ¹³C NMR spectrum included:

· Glucose C-1: δC 101.3 

· Glucose C-2: δC 80.5 

· Rhamnose C-1: δC 102.9 

· Rhamnose C-6: δC 17.8 

The two anomeric proton signals, the two sets of sugar-carbon signals, and the rhamnose methyl signal are consistent with the D-glucose and L-rhamnose identified after hydrolysis. COSY and HSQC further establish proton-coupling relationships within the sugar rings and direct ¹H–¹³C correlations between bonded atoms.

Chemical shifts and coupling constants of sugars are influenced by sugar identity, ring form, substitution position, and conformation. A systematic study by Roslund et al. of D-glucopyranose and D-glucosyl-D-glucosides showed that the position of the interglycosidic linkage can alter the corresponding ¹H and ¹³C chemical shifts and proton-coupling parameters [9].

4 HMBC Determines the Sugar–Aglycone and Sugar–Sugar Linkage Positions

Heteronuclear multiple bond correlation (HMBC) spectroscopy provides long-range ¹H–¹³C correlation information. The classical experiment is primarily used to detect heteronuclear connectivities across two or three chemical bonds [8]. Once the ¹H and ¹³C signals have been assigned, an HMBC correlation between a sugar anomeric proton and a specific carbon atom in another structural unit can directly establish a connectivity constraint between the two structural units.

4.1 Glucose Is Linked to C-7 of Herbacetin

The HMBC spectrum of the unknown glycoside showed the following correlation:

Glucose H-1 (δH 5.20) ↔ herbacetin C-7 (δC 151.9), long-range correlation [1].

Glucose H-1 had already been assigned from the sugar-region signals, and δC 151.9 was assigned to C-7 of herbacetin. This long-range correlation establishes a linkage between the glucose anomeric center and the oxygen substituent at C-7 of herbacetin. The sugar–aglycone moiety was therefore identified as: herbacetin 7-O-glucoside.

4.2 Rhamnose Is Linked to C-2 of Glucose

The second key HMBC correlation was:

Rhamnose H-1 (δH 5.29) ↔ glucose C-2 (δC 80.5) [1].

This correlation establishes the linkage between the rhamnose anomeric center and C-2 of glucose. The interglycosidic linkage is therefore:

rhamnosyl-(1→2)-glucosyl

Here, “1” corresponds to the anomeric carbon C-1 of rhamnose, and “2” corresponds to C-2 of glucose. The δC value of 80.5 for glucose C-2 is also consistent with a glycosylated substitution environment at this position.

Together, the two HMBC correlations establish the complete connectivity framework:

rhamnosyl-(1→2)-glucosyl-O-7-herbacetin

Rhamnosyl-(1→2)-glucosyl corresponds to the neohesperidose structure. MS/MS methodological studies have also shown different fragmentation patterns for 1→2 and 1→6 linkages in some flavonoid O-diglycoside systems; in this compound, the specific C-2 linkage is established by the long-range HMBC correlation [1,4,5].

5 Determination of Anomeric Configurations

The D/L configurational series and the α/β anomeric configurations describe different levels of stereochemical information for sugar residues. Monosaccharide analysis had already established that the two constituent sugars were D-glucose and L-rhamnose; the anomeric configurations in the intact glycoside were further determined from their nuclear magnetic resonance characteristics.

5.1 β Configuration of the D-Glucose Residue

The anomeric proton of glucose was observed at:

δH 5.20, d, J = 7.5 Hz

The vicinal three-bond coupling constant ³J₁,₂ between neighboring protons is related to the dihedral angle between H-1 and H-2 and to the conformation of the sugar ring. In the common chair conformation of a D-glucopyranosyl residue, H-1 and H-2 of the β anomer adopt spatial orientations that can give rise to a relatively large vicinal coupling constant. Nuclear magnetic resonance studies of carbohydrates have likewise shown clear relationships between proton-coupling parameters and sugar-ring geometry and conformation [9,10].

Together with the established identity of D-glucose and the complete spectroscopic assignment of the sugar region, the original study assigned the anomeric signal with J = 7.5 Hz to:

β-D-glucopyranosyl [1].

5.2 α Configuration of the L-Rhamnose Residue

The anomeric proton of rhamnose was observed at:

δH 5.29, d, J = 1.5 Hz

Monosaccharide analysis had already identified this residue as L-rhamnose. On the basis of the complete sugar-region signal assignments for this compound, the original study used the coupling characteristics of the anomeric proton to identify the rhamnose residue as:

α-L-rhamnopyranosyl [1].

The complete stereochemistry and linkage relationship of the two sugar residues can therefore be expressed as:

α-L-rhamnopyranosyl-(1→2)-β-D-glucopyranosyl:

i.e., a neohesperidosyl group.

Interpretation of anomeric coupling constants is related to the specific sugar identity, ring form, and conformation. Different sugars have their own coupling networks and conformational characteristics, and assignment of the anomeric configuration is based on the established sugar identity together with complete nuclear magnetic resonance signal assignments.

6 Correspondence Between Structural Features and Experimental Evidence

The structural information corresponding to the experimental results is summarized below:

Summary

The structure of an unknown glycoside is defined collectively by its molecular composition, constituent units, atomic connectivities, and stereochemical configuration. High-resolution mass spectrometry constrains the elemental composition; hydrolysis and monosaccharide analysis determine the aglycone and constituent sugars; one- and two-dimensional nuclear magnetic resonance spectroscopy establish signal assignments for the individual structural units; HMBC determines the linkage positions between the sugar and aglycone and between sugars; and analysis of monosaccharide absolute configuration together with the nuclear magnetic resonance characteristics of the anomeric centers further establishes the stereochemistry of the sugar residues.

Each element of the complete structure is supported by experimental evidence at the corresponding structural level. Once the molecular formula, aglycone, sugar composition, linkage positions, and anomeric configurations are mutually consistent, the complete structure of the unknown glycoside can be established.

7 Representative Chemical Related to the Determination of Molecular Formula, Sugar Composition, Sugar-Chain Linkages, and Glycosidic-Bond Configuration of Unknown Glycosides

Table 1 Structural Comparison of Neohesperidoside-Type and Rutinoside-Type Flavanones

Table 2 Structural Comparison of Flavone and Flavonol Aglycones and Glycoside Linkage Positions

Table 3 Reference Compounds for Constituent Sugars and Disaccharide Structures

Table 4 Reagents for Acid Hydrolysis, Sugar Absolute-Configuration Analysis, NMR, and Chromatographic/Mass-Spectrometric Analysis

Note: The products listed above are representative Aladdin products related to scientific research. Specific applications should be determined according to the product specifications, batch-specific COA, and the intended reaction or evaluation system. Additional information on product specifications, grades, and COAs can be searched on the Aladdin website using the “product name/CAS/catalog number.”

References

[1] Amakura Y, Yoshimura M, Yamakami S, Yoshida T, Wakana D, Hyuga M, Hyuga S, Hanawa T, Goda Y. Characterization of Phenolic Constituents from Ephedra Herb Extract. Molecules, 2013, 18(5): 5326–5334. DOI: 10.3390/molecules18055326.

[2] Kind T, Fiehn O. Metabolomic database annotations via query of elemental compositions: Mass accuracy is insufficient even at less than 1 ppm. BMC Bioinformatics, 2006, 7: 234. DOI: 10.1186/1471-2105-7-234.

[3] Kind T, Fiehn O. Seven Golden Rules for heuristic filtering of molecular formulas obtained by accurate mass spectrometry. BMC Bioinformatics, 2007, 8: 105. DOI: 10.1186/1471-2105-8-105.

[4] Ma YL, Cuyckens F, Van den Heuvel H, Claeys M. Mass spectrometric methods for the characterisation and differentiation of isomeric O-diglycosyl flavonoids. Phytochemical Analysis, 2001, 12(3): 159–165. DOI: 10.1002/pca.573.

[5] Cuyckens F, Rozenberg R, de Hoffmann E, Claeys M. Structure characterization of flavonoid O-diglycosides by positive and negative nano-electrospray ionization ion trap mass spectrometry. Journal of Mass Spectrometry, 2001, 36(11): 1203–1210. DOI: 10.1002/jms.224.

[6] Shi P, He Q, Song Y, Qu H, Cheng Y. Characterization and identification of isomeric flavonoid O-diglycosides from genus Citrus in negative electrospray ionization by ion trap mass spectrometry and time-of-flight mass spectrometry. Analytica Chimica Acta, 2007, 598(1): 110–118. DOI: 10.1016/j.aca.2007.07.027.

[7] Tanaka T, Nakashima T, Ueda T, Tomii K, Kouno I. Facile Discrimination of Aldose Enantiomers by Reversed-Phase HPLC. Chemical and Pharmaceutical Bulletin, 2007, 55(6): 899–901. DOI: 10.1248/cpb.55.899.

[8] Bax A, Summers MF. Proton and carbon-13 assignments from sensitivity-enhanced detection of heteronuclear multiple-bond connectivity by 2D multiple quantum NMR. Journal of the American Chemical Society, 1986, 108(8): 2093–2094. DOI: 10.1021/ja00268a061.

[9] Roslund MU, Tähtinen P, Niemitz M, Sjöholm R. Complete assignments of the ¹H and ¹³C chemical shifts and JH,H coupling constants in NMR spectra of D-glucopyranose and all D-glucopyranosyl-D-glucopyranosides. Carbohydrate Research, 2008, 343(1): 101–112. DOI: 10.1016/j.carres.2007.10.008.

[10] Tafazzoli M, Ghiasi M. New Karplus equations for ²JHH, ³JHH, ²JCH, ³JCH, ³JCOCH, ³JCSCH, and ³JCCCH in some aldohexopyranoside derivatives as determined using NMR spectroscopy and density functional theory calculations. Carbohydrate Research, 2007, 342(14): 2086–2096. DOI: 10.1016/j.carres.2007.05.032.

For more related articles, see below:

Complete Guide to qNMR Standards: Principles, Selection Criteria, and Widely Used Internal Standards

Should my NMR solvents be handled in a special way?

Structural experiments for the determination of ethylbenzene by nuclear magnetic resonance spectroscopy

nuclear magnetic resonance hydrogen spectroscopy (NMRH) experiment

nuclear magnetic resonance carbon spectroscopy (NMRCS) experiment

Protein identification using nanoscale liquid chromatography-tandem mass spectrometry

Nanoscale two-dimensional liquid chromatography tandem mass spectrometry of rice leaf and root proteins

UHPLC Grade — Suitable for Mass Spectrometry (MS)

Mechanisms by Which Flavonoid Natural Products Regulate the Oxidative Stress-Inflammation Axis

Activity Screening and Activity Tracking Experiments of Traditional Chinese Medicines

Glucosidases: Classification Framework, Catalytic Mechanisms, and Application Essentials

Biological Characteristics and Application Value of the Aminoglycoside Antibiotic Hygromycin B

n-Dodecyl-β-D-maltoside (DDM): Structure, Properties, and Applications as a Non-ionic Surfactant

Molecular Basis of Glycosylation Modification and Functional Release in Natural Products

Will glycosylation affect the biological activity of recombinant protein?

An Introduction to Research-Grade Carbohydrates: Decoding the “Glycan Language” and a Product-Selection Guide (with Tables 1–3)

Plant proteomics and glycosylation experiments

β-1,4-Galactosyltransferase 1: A Precise Glycosyltransferase in Glycan Biosynthesis

Cell Surface Glycan Modification Research: Glycosyltransferases, Glycosidases, and Glycocalyx Functional Analysis

Glycosylation in Biopharmaceuticals